Age | Commit message (Collapse) | Author | Files | Lines |
|
Change-Id: I48ec272fb150c09221142b8d73830d7b6741d0b5
Signed-off-by: Kai Ji <kai.ji@intel.com>
(cherry picked from commit 4a97d3ebadb6c1e30bf868e1eb19ae309d116956)
|
|
Change-Id: Idef14bef816a04ca43ab9aced4b1441ef818e066
Signed-off-by: xinfengx <xinfengx.zhao@intel.com>
|
|
Signed-off-by: Peter Mikus <peter.mikus@protonmail.ch>
Change-Id: I720b7c96a6ac9f328aff57437e51254364604911
|
|
Signed-off-by: Peter Mikus <peter.mikus@protonmail.ch>
Change-Id: I7f640aa28ecf3a60696676b0cf8afb7a080e1b12
|
|
Signed-off-by: Peter Mikus <peter.mikus@protonmail.ch>
Change-Id: I84d9cf2892423e62b116b07982d3e87120b66271
|
|
Change-Id: I8c27b5956c0c26bb2de48321675e285ed961412f
Signed-off-by: xinfengx <xinfengx.zhao@intel.com>
|
|
Signed-off-by: Peter Mikus <peter.mikus@protonmail.ch>
Change-Id: I69ee305b25b9e05d2305cb82aba71ef6418780ab
|
|
No effect on NDRPDR results, just different result packaging
between the MLRsearch library and the rest of CSIT.
- PyPI metadata still to be updated in a separate Change.
Change-Id: I547134da189d1d7761594e92f36cc7c1c232ee32
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Hopefully, with CSIT config values, PDR lower than NDR will not happen.
+ Bump duration_sum default to an odd number,
so users are not surprised by not seeing standard median behavior.
For CSIT this should not matter, overheads hide ties
and number of trials (at least for STL) should stay the same.
Change-Id: Id7130f978c31e71227499612424007c473bcfac2
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Replaces MLRv2, suitable for "big bang" upgrade across CSIT.
PyPI metadata updated only partially (full edits will come separately).
Pylint wants less complexity, but the differences are only minor.
+ Use the same (new CSIT) defaults everywhere, also in Python library.
+ Update also PLRsearch to use the new result class.
+ Make upper bound optional in UTI.
+ Fix ASTF approximate duration detection.
+ Do not keep approximated_receive_rate (for MRR) in result structure.
Change-Id: I03406f32d5c93f56b527cb3f93791b61955dfd74
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: Id646ab09ae51d3153446f1ae334f62ba7710be14
|
|
Change-Id: I4277f23d94fa191890041bb43e2fcd4eb5e42019
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I8945daffd500f689b3eb630f1ce7e1dc4f05aca8
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I4db3fc1b70469a966e51afe3bfa93cb4bcffa654
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I8d168979ab9722e0056e97edb228f29cbfa245e3
|
|
Remove 4 tunnel flow cache test cases, as those don't benefit from flow
cache much.
Add mrr ipsec test cases for the XL710 Intel NIC to increase coverage.
Don't add ndrpdr for XL710, only add commented out counterparts to
align with the mrr list.
Also modify the existing flow cache configuration by adding the fast
path config as well. These are independent features, but the fast path
code added this dependency.
Change-Id: Ide88fd8c1abe96c577a86925dcc2dc4e63f897d4
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
|
|
Previously, the Set up LDP or VCL Nginx on DUT node keyword
was wrongly documenting its "dut" argument as type string,
while using the value as a node dict instead.
As most of the code already used direct dut1 variables,
this patch eliminates the dut argument altogether
and puts DUT1 into the keyword name.
Change-Id: Ie29afc175b98c1b1cb554438da9178738662d108
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I6768f65551deb335ba7b9fc07bca164bca9381fe
|
|
Change-Id: Iac742f0571b11e06662af6bf1b680d0b4e196ddd
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: Idc91da55ab0b5c6b8eb53e8b4ce4da18a1ae891c
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: Iafae1a1f538b928d3bb80bf9610a0bf7de646f8a
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I28af64b7d39a97bbb84f9d987a3f5d3a2d02701b
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I782b87190dbee6e0a12c97f616b80539cd6614bd
|
|
Add ipsec reassembly test cases to bring in line with 3n-icx.
Add ipsec policy test cases with flow cache enabled for both inbound and
outbound traffic.
Remove GTPU hardware tests as the NIC doesn't support GTPU offload RX.
Change-Id: I7b18914f983d5974b390f995ec638a64b9bbccbb
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
|
|
QAT_4xxx is integrated QAT device on SPR platform.
Signed-off-by: xinfeng zhao <xinfengx.zhao@intel.com>
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
Change-Id: If1d4b3d72a2c66855ac8c1c42516f0cb78bdfab2
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I85de04e4a0ed1a642c40c8ca6840cbc1977b10be
|
|
Change-Id: I6ec27d787b59d4bc13b8c04f4217302cc3021581
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I198fe62ddba8910c6bd8256739cc2503b87dfc75
|
|
In tunneling scenarios, VPP needs to reassembly packet fragments
before decapsulation. Conveniently, VPP also fragments the packets
after encapsulation (at least for some encapsulation protocols)
if they do not fit into MTU.
So this change adds few suites, adapted from existing 3-node ones
which additionally lower MTU on DUT1-DUT2 link.
The intended packet size is 1518B. 64B packets will fit,
and most of IMIX packets will also fit without fragmentation.
VPP refuses to reassembly 9000B packets as they have too many fragments.
+ Add a keyword so suites need only one additional line of code.
+ Add new test tag REASSEMBLY for the added suites.
+ Tell autogen the one soak test case for reaseembly should be 1518B.
- It is possible to increase fragment limit for 9000B, not done yet.
- Ipsec policy suites are added but they should not be used.
- VPP does fragmentation twice (before and after encapsulation).
- VPP does not perform reassembly (only last fragment is decrypted).
- TRex does not realize the packet from VPP is only a fragment.
- The result is false pass with unfairly good performance.
+ Suites still included, to simplify verifying VPP fixes for the above.
Change-Id: If33c60d767fea161d9e4ffabb8ded3d81c8f39ed
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
When changing MTU on a running VPP, the interface has to be down.
- Other plugins (rdma, avf, af_xdp) need vastly different logic,
so support for them will be added later.
+ Mlx5-core does not need to set MTU on Linux interface.
+ MTU setting now does not happen at final setting path up,
it happens in driver initialization layer instead
E.g. AVF tests will not attempt to change MTU.
+ MTU edit removed from some non-hardware interfaces (including memif)
e.g. bond interfaces. MTU on parent hw interface seems to be enough.
+ The non-jumbo MTU value used is 1800,
so 1518B tests with additional encapsulation can still work.
+ When VPP MTU setting fails, the failure is now propagated.
Previously, the failure was just logged and ignored,
but now there is no reason to hide it.
Ticket: CSIT-1797
Change-Id: I3b853f1faf90001d544cbbb87b2affbb882ffba0
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I45760143ae0ccb0cc0edf0931ad47f5cf48ec968
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: Icb7b6124dcba7bb57c18ceb91120284f6fe02c2e
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I976f96410b06b4f4d7500b03b1f5dd54486bbf42
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I11896900a73f97194a2935202343dc28756cbb28
|
|
This will greatly improve the performance of IPsec policy cases
Signed-off-by: xinfeng zhao <xinfengx.zhao@intel.com>
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
Change-Id: Ib42b5650791be7fcc8649873eaf12616aeba5aec
|
|
When call create af_xdp interface api, it will load default xdp program
(xsk_def_prog.o) on the NIC interface, the program will be unloaded
when call delete af_xdp interface api, but current csit config did not
call delete api for af_xdp test cases, that will cause af_xdp test case
fails to run on Intel E810 NIC when the interface already have xdp program
attached, so this patch is to fix this issue by to detach any attached
XDP/BPF program on the interface before af_xdp test case run.
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
Change-Id: I1a557b4c4a7e88c5715cec771a001df29ac4ae6f
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I0f5f1000f55110455e86db4c461fba3c87fb198e
|
|
- convert hoststack tests to use app-api-socket instead
of vpp-api-socket
Change-Id: Ibcd87f3624e33168ed3afcb30267262efc927e7c
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
|
|
* remove evt_qs_memfd_seg and evt_qs_seg_size as they are obsolete
* use app socket api instead of vpp's api
* apply rx and tx descriptors constraints
* increase event queue size for nginx tests
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
Change-Id: Ie7ac12b8f587e0b1962884127e8eb68a4e588aee
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I16469204c9bb9d521f0ca01bc3ed21ecba2d8824
|
|
In order to compare performance number with existed routing based ipip
interface mode 40k tunnels ipsec test.
Signed-off-by: xinfeng zhao <xinfengx.zhao@intel.com>
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
Change-Id: I94816a8c18f08b539114d6ca9d4e1218a2b6ea5f
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I23f2ab678e6666a1423620fa373261d822030bc8
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I1c13afd8cb377c76d65f7099b9652b96613ad087
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: Ibfab25ee82eaa207987f4070cf2386ea9d0781cd
|
|
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: Ibb8a59fd4e08bc736494c7ab18e689a55f7a98dc
|
|
- Due to divergence from original design path the RAW was never
consumed. It adds too much code complexity and requires processing
on both storage and compute. Removing entirely to make modeling
efficient.
- log (apparently SSH) section will never be consumed in the way it is
coded in model. This section is also not part of model schema itself
due to the point above.
- Introducing telemetry section that is going to carry telemetry
items required for CDash.
Signed-off-by: pmikus <peter.mikus@protonmail.ch>
Change-Id: I7e0256c6c9715de8ee559eed29dce96329aac97d
|
|
Signed-off-by: Viliam Luc <vluc@cisco.com>
Change-Id: Icd8a5e892bb2c31c01851f731482c692cea4d70a
|
|
Also add 10k tunnels IPSec policy mode test with spd fast path enabled.
spd fast path is to reduce policy adding time, refer to patchset [1] & [2]
for details.
[1] https://gerrit.fd.io/r/c/vpp/+/36095
[2] https://gerrit.fd.io/r/c/vpp/+/36097
Signed-off-by: xinfeng zhao <xinfengx.zhao@intel.com>
Signed-off-by: yulong pei <yulong.pei@intel.com>
Change-Id: I211b38a60d273c46e68a5b5f712090037977c6e6
|
|
Add VPP wireguard async mode test suite to use QAT device for crypto.
Also change keyword ipsechw to cryptohw in suite_setup.robot since currently
crypto device is not only used by IPSec.
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
Signed-off-by: xinfeng zhao <xinfengx.zhao@intel.com>
Change-Id: Ibdadb3b09c04b7181415ffd4a248abddc6289075
|
|
Specifically on 2n-icx, soak tests are suffering
from the first search trial (at half of max rate)
reporting unsent packets, thus polluting the results, see CSIT-1867.
This adds a plain trial (no stats, result ignored), fixing the issue.
To keep the structure similar to NDRPDR tests,
stats trial is done after the search, at the discovered critical rate.
Ticket: CSIT-1867
Change-Id: I12ff6f58c4eb174594e667bfa86ba1585164b8ef
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|